A method for regional estimation of climate change exposure of coastal infrastructure: Case of USVI and the influence of digital elevation models on assessments

This Article is brought to you for free and open access by the Marine Affairs at DigitalCommons@URI. It has been accepted for inclusion in Marine Affairs Faculty Publications by an authorized administrator of DigitalCommons@URI. For more information, please contact digitalcommons@etal.uri.edu. Citation/Publisher Attribution Bove, G., Becker, A., Sweeney, B., Vousdoukas, M., & Kulp, S. (2020). A method for regional estimation of climate change exposure of coastal infrastructure: Case of USVI and the influence of digital elevation models on assessments. Science of the Total Environment, 710, 136162. doi: 10.1016/ j.scitotenv.2019.136162 Available at: https://doi.org/10.1016/j.scitotenv.2019.136162 Follow this and additional works at: https://digitalcommons.uri.edu/maf_facpubs


Introduction 25
Hydrologic models of flooding are sensitive to vertical error and grid size of the underlying the use of lower quality, widely available elevation data in flood models is therefore critical in 39 climate change planning (Gesch 2018). This is particularly important as a uniform data standard 40 is needed for planning at larger scales (e.g., regional) and/or in economically developing countries 41 where high quality data are often not available and the impacts of large storms can be devastating. studies. Although, of these, SRTM offers the best vertical accuracy (Wang, Yang et al. 2012, 46 Gesch 2018) at high horizontal resolution (30m), the data suffer from random noise, voids, striping 47 and other errors that impact accuracy (Falorni, Teles   1.1 Motivation -Coastal infrastructure is at risk, but difficult to assess risk at the regional scale 68 The Low Elevation Coastal Zone (LECZ) (less than 10 meters above sea level) contains   global scale of the risk to coastal infrastructure makes it highly unlikely that resource-constrained 89 SIDS will be able to adapt at a pace adequate to match the threat, even with assistance from number of sites in need of evaluation at a regional scale. Other methods take national, regional, or  The remainder of this paper presents the data components required to efficiently quantify 116 exposure to flooding from storms and sea level rise for critical coastal infrastructure at the 117 individual facility level that is applicable on a regional scale. The method proceeds with identifying 118 critical coastal facilities, creating geospatial data of those facilities, and then applying a dynamic 119 storm model to determine exposure to flooding. Two DEMs -SRTM and a more recent derived 120 product, CoastalDEM v1.1 (Kulp and Strauss 2018) are tested to assess their suitability for a 121 regional level evaluation to be carried out in a subsequent phase of the research.       (Tables 1,2), and 110 features (e.g., parcels, clarifiers, 227 transformers) and 15 roads for utilities (Table 3). A large portion of features (building footprints, 228 tanks, parking lots) of cruise/passenger terminals and marina infrastructure, and 25/32 primary 229 access roads for these facilities were flooded. Utilities related infrastructure tended to be located 230 farther inland than transport infrastructure and fared better overall.  (Table 1).

279
Using readily available data to efficiently identify storm surge exposure over a wide 280 geographic area, this study presents a methodology that bridges a gap between large-scale global 281 or national studies and single-facility case assessments for critical coastal infrastructure. Applying 282 the methodology to the USVI using LIDAR elevation data we found 51% of coastal transportation 283 13 and utilities infrastructure could be exposed to coastal flooding in the coming decades. The same 284 assessment method using CoastalDEM identified 27% of those facilities exposed to coastal 285 flooding, and SRTM matched only 6%. Although the important role topographic data quality and 286 hydraulic model selection play in inundation map accuracy is well established, to our knowledge, 287 this is the first study comparing variations in coastal flood exposure assessment outcomes for 288 coastal infrastructure based on DEMs.

289
There are two primary components that influence exposure estimates, storm model, and 290 digital elevation model, and the impact that each of these have on outcomes can be substantial. in the present study could bring into play many coastal assets that may not be assessed as

355
To our knowledge, this is the first study to assess exposure of critical coastal 356 infrastructure assets that incorporates a method for national or regional scales with specificity to 357 rank facilities by exposure. Although SRTM based DEMs introduce significant error into the 358 assessment, that error does not preclude ranking facilities to efficiently direct resources for 359 further study to protect critical components of local, national and regional economies from 360 climate-related disasters. All coastal infrastructure is vulnerable to the effects of climate change, 361 but not all is equally so and not all will undergo fortification needed to withstand likely impacts.

362
In providing a model that does not require extensive data processing, this method is accessible to 363 analyze infrastructure over broad spatial scales.  There is an urgent need for increased quantity and quality of information on coastal flood risk, 371 but studies should proceed with caution, considering; error associated with the underlying 372 elevation data, error in the approaches used in assessments, and the potential setbacks to progress 373 in climate mitigation when these factors are not carefully considered. This method is not targeted 374 directly at providing informed policy decisions, but as a valuable component towards efficiently 375 achieving that aim.    Electric power substation on St. John, USVI, extreme sea-level and storm surge model tested with SRTM and CoastalDEM elevation with Hit/Miss/False analyses using LIDAR elevation data